JP7599504B2 - Exercise improvement instruction device, exercise improvement instruction method, and exercise improvement instruction program - Google Patents

Description

The present invention relates to a technique for identifying problems for improving athletic performance.

In recent years, mobile communication terminals such as smartphones have become lighter, and the development of so-called wearable devices, such as smart watches, has been active. These devices can be worn while exercising, such as running, and can measure athletic movements using built-in acceleration sensors. There are also services and applications that provide guidance for improving athletic movements based on the results of these measurements.

JP 2014-124448 A

In recent years, as devices have become smarter, the number of built-in sensors has increased, and each sensor has become more sophisticated, allowing a large amount of measurement data to be obtained from a single exercise session. By analyzing such measurement data, it is possible to extract many exercise movement issues from various perspectives, but it is difficult to identify issues that can lead to effective exercise movement improvements.

The present invention has been made in consideration of these circumstances, and its purpose is to provide an exercise improvement instruction device that can identify issues that can lead to effective improvement of exercise movements.

In order to solve the above problems, an exercise improvement instruction device according to one embodiment of the present invention includes a hierarchical index memory unit that stores hierarchical indexes for improving exercise movements, and a problem index identification unit that evaluates the hierarchical indexes based on measurement data during exercise and identifies problem indexes.

In this embodiment, the hierarchical indices are evaluated based on the measurement data during exercise, and problem indices are identified. By referring to the relationships between the indices represented in the hierarchical structure, problem indices that will lead to effective improvements in exercise movements can be identified.

Another aspect of the exercise improvement instruction method of the present invention includes a hierarchical index storage step for storing hierarchical indexes for improving exercise movements, and a problem index identification step for evaluating the hierarchical indexes based on measurement data during exercise and identifying problem indexes.

Another aspect of the exercise improvement instruction program of the present invention causes a computer to execute a hierarchical index storage step for storing hierarchical indexes for improving exercise movements, and a problem index identification step for evaluating the hierarchical indexes based on measurement data during exercise and identifying problem indexes.

In addition, any combination of the above components, and any conversion of the expression of the present invention between a method, device, system, recording medium, computer program, etc., are also valid aspects of the present invention.

The present invention makes it possible to identify issues that can lead to effective improvements in athletic performance.

1 is an overall configuration diagram of a system including an exercise improvement instruction device according to an embodiment; 4 is a diagram illustrating an example of a hierarchical structure of indices stored in a hierarchical index storage unit; FIG. FIG. 11 is a diagram showing a process flow of the exercise improvement guidance device performing exercise improvement guidance using a hierarchical structure. 13 is a diagram showing a process flow of the exercise improvement guidance device providing exercise improvement guidance when the user has performed another run. FIG. FIG. 13 illustrates another example of a hierarchical structure in the form of a directed graph.

In this embodiment, as an example, guidance is given to improve athletic movements such as running form and use of force. Indicators for improving athletic movements are organized into a hierarchy, with the highest hierarchy consisting of "landing impact," "vertical movement," "braking force," and the like. These indicators in the highest hierarchy are gradually subdivided toward lower hierarchies, with the lowest hierarchy consisting of indicators that can be improved individually, such as "ground contact position," "ground contact angle," and "trunk angle." By searching this hierarchical structure of indicators from higher hierarchies to lower hierarchies while referring to the measurement data obtained during running, it is possible to identify problem indicators that should be improved as a priority.

Figure 1 is an overall configuration diagram of a system including an exercise improvement instruction device 100 according to an embodiment. The exercise improvement instruction device 100 generates instruction information for improving the running exercise movement based on measurement data measured by a measurement device 20 while the user is running. The generated instruction information is displayed on a display device 30 used by the user.

The measuring device 20 is, for example, a wearable device such as a smart watch that the user can wear while running, or a smartphone, and measures the exercise movements while running using a built-in sensor. However, in this embodiment, the measuring device 20 is not limited to these, and any device that has a function for measuring data while running and a minimum data transmission function for transmitting the measurement data to the exercise improvement instruction device 100 may be used. For example, a camera (imaging device) that captures an image of the user while running may be used as the measuring device 20. In this case, the data captured by the camera is supplied to the exercise improvement instruction device 100 as measurement data.

In this way, various devices can be used as the measuring device 20, so that the exercise improvement guidance device 100 can obtain various measurement data regarding the user's exercise movements while running.

For example, when a wearable device or smartphone is used as the measuring device 20 as described above, basic physical quantities related to the user's position and movement can be obtained as measurement data from acceleration sensors, angular velocity sensors, position sensors (GPS, etc.), magnetic sensors, etc. built into these devices. By appropriately combining and analyzing these measurement data, not only basic information such as the user's position, speed, and acceleration while running can be obtained, but also information regarding the details of the user's athletic movements, such as the user's posture while running, the strength and angle of the push-off, and the position and angle of the ground contact. Information regarding the running course, such as the distance, altitude, and gradient, can also be obtained.

Furthermore, if the measuring device 20 is equipped with sensors that measure the external environment, such as an ambient light sensor that measures brightness, a temperature sensor, and a humidity sensor, the exercise improvement guidance device 100 can provide appropriate exercise improvement guidance taking into account the external environment during running. Also, if a wearable device capable of measuring biological signals such as heart rate, which has been actively developed in recent years, is used as the measuring device 20, the exercise improvement guidance device 100 can provide appropriate exercise improvement guidance taking into account the user's physical condition.

The measuring device 20 does not need to be worn by the user while exercising, but may be used to measure the user from the surroundings while exercising. A typical example is the above-mentioned camera photography, but this is not limiting. For example, when a user runs in a limited area indoors using a treadmill or the like, a measuring device 20 that is difficult to carry can be used, dramatically increasing the types of measurement data that can be used.

The various measuring devices 20 described above may be used alone or in combination. For example, a user may wear a wearable device as a first measuring device 20 to perform measurements, and a camera as a second measuring device 20 may photograph the user. The exercise improvement instruction device 100 can analyze the user's exercise movements from multiple angles based on the measurement data from such multiple measuring devices 20, and can generate instruction information for effective improvement.

The display device 30 is a device that displays instruction information for improving exercise movements generated by the exercise improvement instruction device 100 based on the measurement data from the measurement device 20. For example, when a smartphone is used as the measurement device 20, it also functions as the display device 30 that displays the instruction information. When a measurement device 20 such as a camera that does not have the function of displaying instruction information is used, another device owned by the user, such as a smartphone, tablet, or personal computer, is used as the display device 30. Note that the means for displaying the instruction information for improving exercise movements generated by the exercise improvement instruction device 100 based on the measurement data from the measurement device 20 is not limited to display by the display device 30. For example, the above instruction information may be configured to be displayed to the user by outputting audio from an audio device.

The exercise improvement instruction device 100 is configured on a server capable of communicating with the measuring device 20 and the display device 30 via a wide area communication network such as the Internet. The exercise improvement instruction device 100 has a hierarchical index storage unit 110, a problem index identification unit 120, and an instruction information generation unit 130. Note that this embodiment is not limited to this, and the exercise improvement starting device 100 may be configured to be able to communicate with the measuring device 20 and the display device 30 via a local communication network such as a LAN. In addition, if the measuring device 20 or the display device 30 is an information processing device such as a smartphone, tablet, or personal computer, the function of the exercise improvement device 100 can also be implemented as application software that operates on these information processing devices. In addition, the data measured by the measuring device 20 may be stored in a storage medium such as a USB memory, and the exercise improvement instruction in this embodiment may be performed by reading it into a personal computer or the like having the function of the exercise improvement instruction device 100.

The hierarchical index storage unit 110 stores hierarchical indexes for improving exercise movements such as running form and how to use force. The hierarchical structure of the indexes stored by the hierarchical index storage unit 110 is configured during the initial setup of the system. In order to provide consistent exercise improvement instruction, the same hierarchical structure configured there can be used for multiple exercise sessions without updating for a certain period of time. The hierarchical structure may be updated by adding new indexes or changing or deleting existing indexes. In addition, teacher data showing the correlation between the performance of the exercise movement and each index may be obtained, and a learning model may be generated by machine learning that inputs the performance of the exercise movement and outputs the corresponding index based on the teacher data. According to such a learning model, it is possible to prioritize indexes that lead to effective improvement of the exercise movement based on the actual performance of the exercise movement. Furthermore, it is also possible to automatically generate a hierarchical structure of the indexes by autonomously analyzing the terms used in each index and their biomechanical meanings.

The problem index identification unit 120 identifies problem indexes by referring to the measurement data during running by the measurement device 20 based on the hierarchical structure of the indexes stored by the hierarchical index storage unit 110. Specifically, the problem index identification unit 120 evaluates the hierarchical indexes based on the measurement data during running, and identifies problem indexes.

The problem index identification unit 120 searches for problem indexes from the upper level to the lower level of the hierarchical structure. At that time, for multiple indexes at the same level in the hierarchical structure, the one with the largest deviation between the specified reference data and the measured data is selected, and problem indexes are searched for from there, moving further down the hierarchy. The problem index identification unit 120 then identifies a problem index at the lowest level of the hierarchical structure. The embodiment is not limited to one in which the search begins at the top level of the hierarchical structure. For example, the user may specify an arbitrary index that he or she wishes to improve in advance, and search for problem indexes from the level of that index towards the lower levels.

The training information generation unit 130 generates training information for improving the problem indicators identified by the problem indicator identification unit 120. The generated training information is transmitted to the display device 30 and displayed on the display screen.

Figure 2 shows an example of a hierarchical structure of indices stored by the hierarchical index storage unit 110. This hierarchical structure is composed of four layers, from the first layer L1, which is the top layer, to the fourth layer L4, which is the bottom layer. In the first layer L1, "landing impact," "vertical movement," and "braking force," which are main indices for evaluating running movements, are arranged as examples.

In the second layer L2, which is directly below the first layer L1, there are indices that are subdivisions of each of the indices in the first layer L1. For example, the "vertical movement" in the first layer L1 is subdivision into two indices, "vertical movement at touchdown" and "vertical movement during flight" in the second layer L2. That is, when there is a problem with the "vertical movement" in the first layer L1, it is roughly divided into a case where there is a problem with the vertical movement at touchdown and a case where there is a problem with the vertical movement during flight, so that an index corresponding to each case is provided in the second layer L2 so that the cause of the problem can be specifically identified. The other indices in the first layer L1, "landing impact" and "braking force", are also similarly subdivision in the second layer L2 and below, but illustrations and explanations are omitted here. When subdivision of an upper hierarchical index into a lower hierarchical index, strict logic is not required, and multiple lower indices may overlap with each other, and a combination of multiple lower indices may not completely reproduce the upper index.

In the third layer L3, which is directly below the second layer L2, there are indices that are subdivisions of each of the indices in the second layer L2. For example, the "vertical movement during flight" in the second layer L2 is subdivision into two indices in the third layer L3: "take-off angle" and "take-off acceleration". In other words, when there is a problem with the "vertical movement during flight" in the second layer L2, it can be broadly divided into cases where there is a problem with the take-off angle at take-off and cases where there is a problem with the take-off acceleration at take-off, so that in the third layer L3, indices corresponding to each case are provided so that the cause of the problem can be specifically identified. The other indices in the second layer L2, "vertical movement at touchdown", are also similarly subdivisions in the third layer L3 and below, but illustrations and explanations are omitted here.

In the fourth layer L4, which is directly below the third layer L3, there are indices that are subdivisions of each of the indices in the third layer L3. For example, the "kicking angle" in the third layer L3 is subdivision into three indices in the fourth layer L4: "ground contact position," "ground contact angle," and "trunk angle." In other words, when there is a problem with the "kicking angle" in the third layer L3, it can be broadly divided into cases where there is a problem with the ground contact position, where there is a problem with the ground contact angle, and where there is a problem with the trunk angle, so that in the fourth layer L4, indices corresponding to each case are provided so that the cause of the problem can be specifically identified. Note that the other indices in the third layer L3, "kicking acceleration," are also similarly subdivisions in the fourth layer L4, but illustration and explanation are omitted here.

In the above-mentioned hierarchical structure, by sequentially evaluating each index from the upper layer to the lower layer while referring to the measurement data obtained during running, it is possible to specifically identify the problem index to be improved. For example, even if there is a problem with the "vertical movement" of the first layer L1, which is the highest layer, it is not possible to provide appropriate exercise improvement guidance unless the specific cause is identified. According to the hierarchical structure of this embodiment, when the problem "vertical movement" in the upper layer is identified, it is possible to search for the specific cause from there toward the lower layers. And, for example, when the "landing position" of the fourth layer L4, which is the lowest layer, is identified as the cause, guidance information for improving it, such as "land one step forward," can be generated.

3 shows a process flow of exercise improvement guidance performed by the exercise improvement guidance device 100 using the above-mentioned hierarchical structure. In the following description, "S" stands for step.
In S10, measurement is performed by the measuring device 20 while the user is exercising.
In S20, measuring device 20 transmits measurement data to exercise improvement guidance device 100. At this time, when the exercise measurement in S10 is performed using multiple measuring devices 20, the measurement data of the multiple measuring devices 20 may be transmitted to exercise improvement guidance device 100. Furthermore, when multiple pieces of data are measured using one measuring device 20, the measurement data of the multiple measuring devices 20 may be transmitted to exercise improvement guidance device 100.

In S30, the exercise improvement instruction device 100 processes the various measurement data received in S20 and converts each index in the hierarchical structure into evaluable measurement data. Many of the indexes in the hierarchical structure cannot be evaluated using the data measured by the measuring device 20 as is, so appropriate calculation processing must be performed to convert them into evaluable data. Various conversion methods are known in the art, so a detailed explanation will be omitted, but for the following indexes described in this flowchart, measurement data for evaluation can be obtained, for example, as follows:

"Up and down movement" of the first layer L1: This can be measured directly using an acceleration sensor, etc.
“Vertical movement during flight” of the second layer L2: It is possible to measure vertical movement by distinguishing between flight and contact using an acceleration sensor or the like.
"Kick-off angle" of the third layer L3: Can be calculated from the measurement data of the acceleration sensor and angular velocity sensor when transitioning from a grounded state to an airborne state.
"Landing position" of the fourth layer L4: The relative position of the landing position to the takeoff position is calculated from measurement data from position sensors etc. at the time of takeoff (landing → airborne) and at the time of landing (airborne → landed).

In S40, the problem index identification unit 120 searches for problem indexes based on the measurement data processed in S30. S40 is made up of S41 to S44, which correspond to the four levels L1 to L4 of the hierarchical structure. By performing the processes in the order of S41, S42, S43, and S44, a search for problem indexes is performed from the higher levels to the lower levels of the hierarchical structure. More specifically, multiple indexes at each level are evaluated based on the measurement data, the most problematic index is selected, and the cause is searched for from there, moving further down the levels.

The search step S41 of the first layer L1 is composed of S411 to S413.
In S411, a plurality of indices in the first layer L1 are evaluated based on the measurement data. Specifically, three indices in the first layer L1, "landing impact", "vertical movement" and "braking force", are evaluated. Each indices has a reference data indicating its normal value set in advance, and the reference data processed for evaluation of each indices in S30 is compared with the reference data to obtain the degree of deviation. Then, based on the magnitude of the deviation, it is evaluated whether or not the indices have a problem. In the following, for the sake of simplicity, the standard for evaluating all indices as having a problem is set to "a deviation of more than 10%", but in reality, different standards can be set for each indices. In addition, the magnitude of the deviation represents the urgency of the indices, that is, the priority in exercise improvement guidance. Therefore, between an index with a deviation of 20% and an index with a deviation of 15%, the former has a higher urgency and is a priority index in exercise improvement guidance.

In S412, it is determined whether or not there was a problematic indicator in S411. That is, it is determined whether or not there was a deviation of more than 10% for the three indicators in the first layer L1, "landing impact," "vertical movement," and "braking force." If the deviation for any indicator is less than 10%, it can be said that all indicators, including their respective sub-indices, are within the normal range, so the problem indicator identification unit 120 does not identify a problem indicator, and the training information generation unit 130 ends the process without generating training information. In such a case, the training information generation unit 130 may generate a message to motivate the user, such as "Keep up the good work."

On the other hand, if there is an index with a deviation of more than 10% in S412, the index with the largest deviation is selected in S413. In the hierarchical structure of Figure 2, for example, if the deviation of "landing impact" is 15%, the deviation of "vertical movement" is 20%, and the deviation of "braking force" is 7%, there are problems with "landing impact" and "vertical movement" with deviations exceeding 10%, but "vertical movement" with the largest deviation has a high urgency and is selected as the priority index. In the subsequent steps, a search for problem indexes is performed from there, moving further down the hierarchy.

The search step S42 of the second layer L2 searches for lower level indicators of the priority indicator "vertical movement" selected in S413, and is composed of S421 to S422.
In S421, multiple indices of "vertical movement at touchdown" and "vertical movement during flight" in the second layer L2, which are sub-indices of the priority index "vertical movement" in the first layer L1, are evaluated based on the measurement data. Specifically, for each index, a predetermined reference data is compared with the measurement data to determine the degree of deviation.

In S422, the index with the largest deviation is selected. For example, if the deviation for "vertical movement at touchdown" is 8% and the deviation for "vertical movement in the air" is 15%, then "vertical movement in the air" has the highest deviation and is therefore selected as the priority index. In the subsequent steps, a search for problem indexes is performed from there, moving down to lower levels.

The search step S43 of the third layer L3 searches for lower level indicators of the priority indicator "vertical movement during flight" selected in S422, and is composed of S431 to S432.
In S431, multiple indicators of the third layer L3, which are sub-indices of the priority indicator of the second layer L2, "airborne vertical movement," such as "kick-off angle" and "kick-off acceleration," are evaluated based on the measurement data. Specifically, for each indicator, the measurement data is compared with predetermined reference data to determine the degree of deviation.

In S432, the index with the largest deviation is selected. For example, if the deviation of "kick-off angle" is 17% and the deviation of "kick-off acceleration" is 8%, the "kick-off angle" with the largest deviation has a high urgency and is selected as the priority index. In the subsequent steps, a search for problem indexes is performed from there, moving down to lower levels.

The search step S44 of the fourth layer L4 searches for lower level indicators of the priority indicator "kick-off angle" selected in S432, and is made up of steps S441 to S442.
In S441, the multiple indices "ground contact position,""ground contact angle," and "trunk angle" of the fourth layer L4, which are sub-indices of the priority index "take-off angle" of the third layer L3, are evaluated based on the measurement data. Specifically, for each index, the measurement data is compared with predetermined reference data to determine the degree of deviation.

In S442, the index with the largest deviation is selected. For example, if the deviation of "ground contact position" is 20%, the deviation of "ground contact angle" is 13%, and the deviation of "trunk angle" is 5%, then the "ground contact position" with the largest deviation has a high degree of urgency and is selected as the priority index. Then, the problem index identification unit 120 identifies the selected priority index "ground contact position" as the problem index.

In S50, the instruction information generating unit 130 generates instruction information such as "Land one step forward" for improving the problem indicator "ground contact position" identified by the problem indicator identifying unit 120. The instruction information generated here is transmitted to the display device 30 and displayed on the display screen.

According to the above processing flow, by referring to the relationships between each indicator represented in a hierarchical structure, it is possible to identify problem indicators that will lead to effective improvement of athletic movements and generate coaching information.

By searching for problem indicators from the higher levels to the lower levels in the hierarchical structure, the problem indicator identification unit 120 can identify major problems in the higher levels and their underlying causes in the lower levels, thereby enabling effective exercise improvement guidance to be provided.

This search for problem indicators is performed sequentially from the first layer L1, which is the top layer, to the fourth layer, which is the bottom layer, and in the search processes S41, S42, S43, and S44 in each layer, priority indicators are identified based on the deviation between the actual measured value and the reference value for each indicator. This makes it possible to identify the indicator with the highest urgency in each layer and identify the indicator that needs improvement with the highest priority as the problem indicator.

By the problem indicator identification unit 120 identifying problem indicators in the fourth layer L4, which is the lowest layer in the hierarchical structure, it is possible to identify problem indicators that represent the root cause of a problem and that are easy to generate specific guidance information for improvement.

Figure 4 shows a process flow of the exercise improvement instruction device 100 performing exercise improvement instruction when the user performs another run after the exercise improvement instruction according to the processing flow of Figure 3 is performed. In this process, when the problem index identification unit 120 identifies an index that has improved to a certain degree compared to the past running, it searches for the problem index among other indexes in the same hierarchy or a higher hierarchy as the index. Also, when the problem index identification unit 120 identifies the same problem index as the past running, the instruction information generation unit 130 generates instruction information different from the past running. In Figure 4, the processes of S10, S20, S30, and S40 are performed as in Figure 3, and in S442, which is the final step of S40, a priority index with a large degree of deviation is selected. In the following, the previous running time is taken as an example of the past running time, but the process of this embodiment can be applied to past running times before the previous time.

In S60, it is determined whether the priority index selected in S442 for the current exercise is the same as the priority index "ground contact position" selected in S442 for the previous exercise. If the priority index is different from that of the previous exercise, the problem index identification unit 120 identifies the priority index as a problem index, and the guidance information generation unit 130 generates guidance information for improving the problem index in S50. For example, if the problem index for the current exercise is "trunk angle," guidance information such as "keep your trunk straight" is generated.

On the other hand, if the priority indicator is the same "ground contact position" as during the previous exercise, in S64 it is determined whether or not there has been a predetermined degree of improvement in that indicator compared to the previous exercise. The criteria for determining whether or not there has been improvement can be set arbitrarily for each indicator, but the deviation degree described above can be used. For example, it can be determined that there has been an improvement if the deviation degree is 5% or more smaller compared to the previous exercise. Since the deviation degree for the "ground contact position" during the previous exercise was 20%, if the deviation degree during the current exercise is smaller than 15%, it is determined that there has been a predetermined degree of improvement.

If it is determined in S64 that there has not been a predetermined level of improvement in the "ground contact position," i.e., if the deviation is still high at 15% or more, the problem indicator identification unit 120 identifies the same "ground contact position" as during the previous exercise as the problem indicator. In this case, the training information generation unit 130 generates training information again in S51 to improve the "ground contact position," but generates training information that is different from the previous time. This allows the user to improve the same problem indicator from different perspectives.

If it is determined in S64 that there has been a predetermined degree of improvement in the "ground contact position," i.e., if the deviation is less than 15%, the search continues in the higher level, the third layer L3. Specifically, in S63, it is determined whether there has been a predetermined degree of improvement in the "take-off angle" (third layer L3), which is a higher level indicator of the "ground contact position" (fourth layer L4), compared to the previous exercise.

If it is determined in S63 that the "kick-off angle" has not improved to a predetermined degree, there is still a problem index that should be improved preferentially in the lower layer, so the process returns to S44 in FIG. 3, and the search continues in the lower layer (fourth layer L4). At this time, since the deviation of the "ground position" is the largest in the fourth layer L4, if S442 in S44 is executed as is, the "ground position" will end up being identified as the problem index. However, since the "ground position" has improved to a predetermined degree compared to the previous exercise (S64), it may be excluded from the candidates for the problem index for the current exercise. In that case, in S442, the index with the largest deviation among the other indices "ground angle" and "trunk angle" excluding the "ground position" will be identified as the problem index. On the other hand, S442 may be executed as is to identify the same "ground position" as in the previous exercise as the problem index. In that case, as explained in relation to S51, it is preferable to generate guidance information different from the previous one.

If it is determined in S63 that there has been a predetermined improvement in the "take-off angle," the search continues in the higher level, the second level L2. Specifically, in S62, it is determined whether there has been a predetermined improvement in the "airborne vertical movement" (second level L2), which is a higher level indicator of the "take-off angle" (third level L3), compared to the previous exercise.

If it is determined in S62 that there has not been a predetermined level of improvement in "vertical movement while airborne," there are still problem indicators in the lower hierarchical level that need to be improved as a priority, so the process returns to S43 in Figure 3, and the search in the lower hierarchical level (third hierarchical level L3) continues.

If it is determined in S62 that there has been a predetermined improvement in "vertical movement while in the air", the search continues by moving to the higher level, the first level L1. Specifically, in S61, it is determined whether there has been a predetermined improvement in "vertical movement" (first level L1), which is a higher level indicator of "vertical movement while in the air" (second level L2), compared to the previous exercise.

If it is determined in S61 that there has not been a predetermined level of improvement in "up and down fluctuations," there are still problem indicators in the lower hierarchy that need to be improved as a priority, so the process returns to S42 in Figure 3, and the search in the lower hierarchy (second layer L2) continues.

If it is determined in S61 that there has been a predetermined improvement in "vertical movement", the process returns to S412 in FIG. 3, and a determination is made as to whether there are any problematic indicators in the first layer L1. At this time, since "vertical movement" has improved to a predetermined degree compared to the previous exercise, it is preferable to exclude it from the judgment target of S412. In that case, in S412, the judgment is made on the other indicators excluding "vertical movement", namely "landing impact" and "braking force".

According to the above process flow, it is possible to efficiently search for problem indicators for the current exercise, taking into account improvements from previous exercises. At this time, the improvement judgment processes S64, S63, S62, and S61, which are executed in sequence from the lower levels to the higher levels of the hierarchical structure, can efficiently identify up to which level improvement has occurred.

The present invention has been described above based on an embodiment. The embodiment is merely an example, and it will be understood by those skilled in the art that various modifications are possible in the combination of each component and each treatment process, and that such modifications are also within the scope of the present invention.

In the above embodiment, the problem index identification unit 120 searches for problem indexes from the upper level to the lower level in the hierarchical structure, but the search method in the hierarchical structure is not limited to this. For example, instead of starting the search for problem indexes from the top level, the search may start from a level specified in advance by the user or the system. It is also possible to search the lowest level from the beginning and identify the index with the largest deviation as the problem index. In this search, indices in levels other than the lowest level are not taken into account, but when providing exercise improvement guidance taking into account improvements from past exercise as described in FIG. 4, indices in each level are taken into account as in S64, S63, S62, and S61.

In the above embodiment, all the indices in the top layer L1 are subdivided down to the indices in the bottom layer L4, and the depth of the hierarchical structure is uniformly four layers. However, the depth of the hierarchical structure may differ depending on the indices. That is, if the top layer is L1 as in the embodiment, there may be an index of depth 1 where there is no hierarchical structure below the first layer L1, an index of depth 2 that is subdivided down to the second layer L2, or an index of depth 3 that is subdivided down to the third layer L3. In this case, there may not be indices corresponding to all layers. For example, an index obtained by subdividing an index in the first layer L1 may not exist in the second layer L2 directly below, and may be further subdivided into an index in the third layer L3. In addition, the top layer does not need to be the first layer L1 for all indices, and the top index may be in the second layer L2, the third layer L3, or the fourth layer L4. When searching hierarchical structures of different depths as described above, it is not necessary to search down to the lowest level, and the user or system can specify the depth to search for each indicator each time. For example, even when searching a hierarchical structure of depth 3, if it is specified to search up to depth 2, the lowest level of depth 3 will not be searched, and problem indicators will be identified in the level of depth 2 above.

In the above embodiment, the problem index is identified in the lowest hierarchical layer of the hierarchical structure, but the problem index may be identified in other layers. For example, the index with the largest deviation across all layers may be identified as the problem index. In this case, it is preferable to prepare guidance information corresponding to all indices regardless of the layer. Also, if guidance information is generated for an index of a certain layer during a past exercise, and as a result, the index of that layer has sufficiently improved during the current exercise, the problem index can be identified in a higher layer. This allows guidance based on an appropriate layer to be provided in accordance with the improvement of the user's exercise movements.

In the above embodiment, when the problem index identification unit 120 searches for a problem index, the index with the largest deviation between the reference data and the measured data in each hierarchy is selected, and the problem index is further searched for toward the lower hierarchy. However, multiple indexes with deviations greater than a preset value may be selected in each hierarchy. In such a case, a search is performed in the lower hierarchy for each selected index, and multiple problem indexes are identified. The guidance information generation unit 130 generates guidance information for each identified problem index and displays it on the display device 30. On the display device 30, the multiple pieces of guidance information may be displayed on a single screen, or the user may switch between multiple screens to check the guidance information for each problem index.

In the processing flow of FIG. 4, if an index in a certain layer has improved to a certain degree compared to a previous exercise, the presence or absence of improvement was determined for the index in the layer above, but the presence or absence of improvement may also be determined for other indexes in the same layer. For example, in FIG. 4, if there is an improvement in the "ground contact position" in the fourth layer L4 in S64, the presence or absence of improvement was determined for the "push-off angle" in the third layer L3 in S63, but the presence or absence of improvement may also be determined for other indexes in the fourth layer L4, the "ground contact angle" and "trunk angle". These indexes in the fourth layer L4 are dependent on the "push-off angle" in the third layer L3, so it makes no difference which layer the improvement is determined in.

In the above embodiment, a tree structure as shown in Figure 2 was exemplified as a hierarchical structure, that is, a structure that branches from a higher hierarchy to a lower hierarchy, but the hierarchical structure of the present invention is not limited to such a tree structure. Figure 5 shows another example of a hierarchical structure in the form of a directed graph. In this figure, vertices A to K of the directed graph represent the indices of each of the hierarchies L1 to L4, and the directed edges connecting each of the indices indicate the direction of search for problem indices. Indicators A, B, and C exist in the first layer L1 of the top hierarchy.

Indicators D and E exist in the second layer L2. Here, indicator D is connected by an edge (arrow) starting from indicators A and B in the first layer L1, respectively, which means that indicator D is evaluated when the deviation of indicators A and B is greater than a predetermined value. Therefore, if the deviation of at least one of indicators A and B is equal to or less than a predetermined value, the search for problem indicators ends in the first layer L1 for indicators A and B. At this time, if the deviations of indicators A and B are both equal to or less than a predetermined value, neither is a problem indicator, and if the deviation of either one is greater than a predetermined value, it is identified as a problem indicator. Although similar explanations will be omitted from here on, a directed edge (arrow) means that if the deviation is greater than a predetermined value, the search proceeds to a lower hierarchical indicator. Unlike the tree structure in FIG. 2, in this example, it can also be considered that the upper indicators A and B are integrated into the lower indicator D.

The third layer L3 has indexes F, G, and H. Index F is specified by an arrow from index D in the second layer L2. Index G is specified by an arrow from index E in the second layer L2 and index C in the first layer L1. Index H is specified by an arrow from index E in the second layer L2.

The fourth layer L4 has indices I, J, and K. Index I is designated by an arrow from indices F and G in the third layer L3. Index J is designated by an arrow from index G in the third layer L3. Index K is designated by an arrow from indices F and H in the third layer L3.

According to the above-mentioned hierarchical structure using a directed graph, it is possible to appropriately describe the relationships between indices that are more complex than those in the tree structure shown in Figure 2, and to effectively identify problem indices. Also, in Figure 5, only arrows pointing from the upper level to the lower level are illustrated, but the process of searching for improvements from past exercise from the lower level to the upper level as described in relation to Figure 4 can be described in Figure 5 by arrows pointing from the lower level indicators to the higher level indicators.

In this embodiment, running has been used as an example of exercise, but the present invention can also be applied to other exercises. For example, the present invention can be applied to training and exercise in various track and field sports, swimming, gyms, walking, road bikes, dancing, and ball games such as soccer.

The functional configuration of each device described in the embodiments can be realized by hardware resources or software resources, or by the cooperation of hardware and software resources. Processors, ROM, RAM, and other LSIs can be used as hardware resources. Programs such as operating systems and applications can be used as software resources.

100: Exercise improvement instruction device, 110: hierarchical index memory unit, 120: problem index identification unit, 130: instruction information generation unit, 20: measurement device, 30: display device.

The present invention relates to an exercise improvement instruction device that identifies issues for improving exercise movements.

Claims (8)

a hierarchical index storage unit for storing hierarchical indexes for improving exercise movements;
a problem index identifying unit that identifies a problem index by evaluating the hierarchical index based on measurement data during exercise;
The hierarchical structure of the index includes at least three levels, and an index other than the lowest level is subdivided into a plurality of indexes in the level immediately below it;
The problem indicator identification unit selects, from among multiple indices at the same level in the hierarchical structure of the indices, those in which the deviation between specified reference data and the measurement data is greater than a preset value, and then searches for the problem indicators in a stepwise manner, moving further down to at least two lower levels. The exercise improvement instruction device according to claim 1 , further comprising an instruction information generating unit that generates instruction information for improving the problem indicator identified by the problem indicator identifying unit. The exercise improvement instruction device according to claim 1 , wherein the problem index identifying unit searches for the problem index from a prespecified hierarchical level of a hierarchical structure of the indices. The exercise improvement instruction device according to claim 1 , wherein the problem index identifying unit identifies the problem index in a lowest hierarchical level of a hierarchical structure of the indices. 5. The exercise improvement guidance device according to claim 1, wherein when the problem index identifying unit identifies an index that has improved to a predetermined degree compared to a previous exercise, the problem index identifying unit searches for the problem index among other indexes in the same hierarchy or a higher hierarchy than the identified index. The exercise improvement instruction device according to claim 2 , wherein when the problem indicator identifying unit identifies the same problem indicator as that during a past exercise, the instruction information generating unit generates the instruction information different from that during a past exercise. A hierarchical index storage step of storing a hierarchical index for improving athletic movements;
and a problem index identifying step of evaluating the hierarchical indexes based on measurement data during exercise and identifying a problem index,
The hierarchical structure of the index includes at least three levels, and an index other than the lowest level is subdivided into a plurality of indexes in the level immediately below it;
The problem indicator identification step is a method for providing exercise improvement guidance in which, for a plurality of indices at the same level in a hierarchical structure of the indices, those in which the degree of deviation between specified reference data and the measurement data is greater than a preset value are selected, and the problem indicator is then searched for in stages, moving further downward to at least two lower levels. A hierarchical index storage step of storing a hierarchical index for improving athletic movements;
a problem index identifying step of evaluating the hierarchical indexes based on measurement data during exercise and identifying a problem index;
The hierarchical structure of the index includes at least three levels, and an index other than the lowest level is subdivided into a plurality of indexes in the level immediately below it;
The problem indicator identification step is an exercise improvement instruction program that selects, for multiple indicators at the same level in the hierarchical structure of the indicators, those in which the deviation between specified reference data and the measurement data is greater than a preset value, and then searches for the problem indicators in a stepwise manner, moving further down to at least two lower levels.

Images